An NCDB study examining disparities in the administration of immunotherapy among advanced cervical cancer patients.

Abstract

5524 Background: The addition of immunotherapy (IO) to standard therapy has demonstrated substantial survival benefits in the upfront treatment of advanced cervical cancer. Our primary objective was to investigate the role of race on the administration of IO among patients with stage IV cervical cancer, following the publication of KEYNOTE-826 in 2021. Secondarily, we aimed to describe additional factors that may be associated with disproportionate administration. Methods: This was a retrospective cohort study utilizing the National Cancer Database (NCDB). We included all patients diagnosed in 2022 with stage IV adenocarcinoma, adenosquamous carcinoma, and squamous cell carcinoma of the cervix. Proportions were used to estimate the administration of IO. Multivariable models adjusted for age, insurance, education, Charlson-Deyo comorbidity scores, geography, distance to treatment center, as well as treatment facility type, location, and volume. Results: There were 937 cases identified, of which 368 (39.3%) received IO and 569 (60.7%) did not. A higher proportion of patients received IO at age < 65 compared to > 65 (74.5% vs 25.5%). Patients with Medicaid received IO at similar rates compared to privately-insured patients, at 33.4% verses 34.5%, respectively. When controlling for all other variables, higher education was associated with a greater likelihood of receiving IO (RR 1.46, CI 1.10 – 1.95, p=0.01). Patients with comorbidity scores > 2 were less likely to receive IO than those with lower scores (RR 0.71, CI 0.52 – 0.98, p=0.038). Proportions of patients receiving IO in urban verses rural communities was also similar, at 33.3% and 39.4%, respectively. There were no significant differences in administration of IO associated with treatment center type (RR 1.05, CI 0.89 – 1.23, p=0.55) or volume (RR 0.95, CI 0.79 – 1.13, p=0.55). Mean distances to treatment centers were similar between patients who did and did not receive IO (29.31 vs 30.0 miles). Hispanic patients received IO at higher rates than non-Hispanic White, non-Hispanic Black, and Asian patients at a rate of 47.2% compared to 37.6%, 38.5%, and 38.1%, respectively. Even after adjusting for demographic and tumor factors, the increased receipt of IO among Hispanic patients could not be explained. Conclusions: Not only were there no racial disparities observed in the administration of IO among racial groups, but there were no disparities recognized among traditionally marginalized groups. Interestingly, the overall administration IO was lower than expected. The population included in KEYNOTE-826 had a PD-L1 combined positivity score of >1 among approximately 88% of participants. This difference in overall IO administration may be explained by PD-L1 positivity, however our study was limited by the lack of this data. Further investigation is warranted to understand trends in IO administration.